With multiple energy sources, diverse energy demands, and heterogeneous socioeconomic factors, energy systems are becoming more and more complex and multifaceted. Therefore, it is also becoming more and more challenging to improve the efficiency, security and sustainability for such complicated systems. To address these challenges, we propose a general optimal dispatch model for integrated energy systems. Two interesting and challenging decisions of any such model is how it takes power curves of equipment into account and how it deals with nonlinearity. We use a Gaussian Process (GP) to estimate the dynamic power curves, and then we linearize the nonlinear program using special ordered sets of type 2 and solve it using CPLEX. To demonstrate the practicality of the proposed approach, we combine real world and simulated scenarios to perform extensive experiments.
Keywords virtual integrated energy system, renewable energy, MILP, optimal dispatch model, dynamic power curve